I came across this circuit (attached) and built it using a couple of 2.2M pcb pots for R2 and R3. It seems to work best at full 2.2M so I'd like to replace the pots with fixed resistors. I can't seem to find any 2.2M but only 2.0M or 2.7M. I'm a simple "follow the recipe" guy so would appreciate advice. Should I use the 2.0M or the 2.7M fixed resistors for R2 and R3. I would like to have the highest clean output I can to drive some long cables.
Thanks.
Thanks.
Those values are completely non-critical. They do set the input impedance, and along with C1, the bass roll off point.
I wouldn't go any higher, though.
Thanks. So no higher than 2.0M?
Interesting because the author actually recommends going as high as 10M or even 22M for what he calls a "more neutral sound" for humbucker pups.
I plan to use it for guitars with single coil and humbucker pups. Is there a good compromise for value of R2 and R3?
Interesting because the author actually recommends going as high as 10M or even 22M for what he calls a "more neutral sound" for humbucker pups.
I plan to use it for guitars with single coil and humbucker pups. Is there a good compromise for value of R2 and R3?
All kinds of parasitic effects will come into play at very high impedances. The highs will start to suffer at some point due to Miller effect.
Certainly you can try going even higher in the resistor values. The bass can be extended by increasing the capacitor instead, though.
Is there some reason why you can't solder two resistors together in series to get the value you want?
Thanks mcgruff. No reason, that is an option too. I have some 1.8M and some 470K (=2.27). Just trying to keep it uncluttered and was curious about whether my best option is to use 2.0M or a 2.7M.
Your opinion?
Your opinion?
I think I'd try some experiments. Resistors are cheap so test a few higher-value ones including the 10M your author suggests.
If single coils & humbuckers work better with different values add a switch option.
If single coils & humbuckers work better with different values add a switch option.
Thanks. That's a good suggestion. However the buffer/splitter is for live performance situations and I'd like to avoid having another switch or button to press so am hoping I could use fixed values for R2/3 that wouldn't overly compromise the tone of either single or hummer pups. As I tend to use the single coil gtr more frequently I guess the best choice would be as close to 2.2M as I can get?
Ive built those buffers a few times into guitars and stomp boxes. They work very well, and not too critical. But its worth knowing that jfets are a bit inconsistent, not only between types but individual examples too. For very best performance, youd like the voltage at the top of r4 (the jfet source) to be about 5v to 5.5v for a 9v supply. This may require raising the value of R2 in comparison to R3. This is very likely with mpf102 and less likelg with j201.You can certainlg try ths values you have and see what works best. I use 2n5457 and find r2 is usually a bit higher than r3.
Last edited:
Since this is a FOLLOWER circuit, you ideally want the SOURCE node to be at roughly 1/2 the 9V supply + Vds ..... 5V is fine....since your probably not running into the rail with this circuit...
The two 2.2 M resistors are in Parallel with respect to AC input impedance... so the author is most likely aiming towards the tradition 1M input imedance used in high-Z musical instrument inputs...
Since 2.2Meg resisitors very common resistors, more so than 2M , I believe that is why the author used them...
One problem with these size resistors and larger values is the Johnson Noise they create... The other issue is the input capacitance of the JFET is non-linear ....In follower configuration you should be OK ... The +9V need to be very well bypassed...Good low ESR cap, with enough uF to pass lowest frequency...or bypass with several types of caps...
I believe the link okcrum provided will help you get the parts you need ordered...
At one time even Radio Shmuck sold 2.2M 1/4W resisistors.... I still have the Archer brand package in the back of the workbench 🙂
If your in a hurry, then use 4 of 1M resistors and make 2M resistors for R2 and R3....
The two 2.2 M resistors are in Parallel with respect to AC input impedance... so the author is most likely aiming towards the tradition 1M input imedance used in high-Z musical instrument inputs...
Since 2.2Meg resisitors very common resistors, more so than 2M , I believe that is why the author used them...
One problem with these size resistors and larger values is the Johnson Noise they create... The other issue is the input capacitance of the JFET is non-linear ....In follower configuration you should be OK ... The +9V need to be very well bypassed...Good low ESR cap, with enough uF to pass lowest frequency...or bypass with several types of caps...
I believe the link okcrum provided will help you get the parts you need ordered...
At one time even Radio Shmuck sold 2.2M 1/4W resisistors.... I still have the Archer brand package in the back of the workbench 🙂
If your in a hurry, then use 4 of 1M resistors and make 2M resistors for R2 and R3....
Last edited:
Thanks. I have some 2.0M so I'll just go ahead and use them. Would there be any problem if I bypassed the buffer/splitter and when straight to the guitar amp while still having the output from the buffer/splitter feed the vocal harmonizer 30 feet away?
Should work OK, but do also try the buffered signal straight to the amp, you might like the extra zing. That is particularly the case if you build a buffer in the guitar, or use very short cable before the buffer.
I should point out a couple of things though:
If you have any decent non true-bypass pedals, such as a Boss tuner, they already have a good buffer in them and you can use that.
If connecting your overall rig across a distance, these circuits will not, on their own, address any issues of ground-loop hum, if you have such problems. (starting point is to bring all equipment in the rig to a common ground, eg back to one mains outlet)
Cerrem - I'm also aware of that theoretical noise where dc passes through high-value resistors. (I believe the reason for it is based on quantum physics!) I've listened for it on circuits like this and honestly cant detect it in practice. Metal oxide resistors are supposed to reduce it. Also, these circuits can be biased in a different way, with a voltage divider made of say, 100k resistors, with one bypassed by a cap to ground to sink any noise, then feed the jfet gate via a 1M resistor from there. But I've not built this.
Last edited:
Thank you. Perhaps I should have started this thread by outlining my root issue which is really getting a solid guitar output signal to the vocal harmonizer which is about 30' away from me in a performance situation AND a good guitar output to the effects. Without the need to drive the vocal harmonizer my guitar output goes straight into a multi-effects pedal (Line 6) that I understand has a 1M input impedance. The output from the pedal goes directly to a vintage (DLXR) tube amp near me and is mic'd to the board. The vocal harmonizer wants to see a guitar level output at its input. (i.e. not post effects). I've tried a passive "Y" splitter, a couple of different DI boxes but I either don't have enough signal level to drive the harmonizer or the DI output seems to "color" the guitar tone in a displeasing manner ("too hot"). I built the buffer/splitter in an attempt to satisfy both the input to the harmonizer and the input to the effects pedal. Perhaps I've gone down the wrong rabbit hole. (I only know enough electronics to be dangerous!)
yep, should work!
Trouble with these threads is that sometimes, we get into more detail than needed! All interesting though.
Trouble with these threads is that sometimes, we get into more detail than needed! All interesting though.
Just one last thing. Almost any jfet will work, but if you let me know the type that you have, I could offer a best guess at how to best wire your resistors. eg, a common type is an MPF102. With those, you would actually leave off the upper resistor R2.
Ok, with MPF102 its worth trying with R2 omitted, or at least much larger (say x2 to x3) than R3.
The key thing is that the ratio of R3/(R2+R3) sets the voltage at the gate. The voltage at the source rises above this according to a parameter Vgs, which is relatively high with that JFet, can be 2.5 to 4V.
The key thing is that the ratio of R3/(R2+R3) sets the voltage at the gate. The voltage at the source rises above this according to a parameter Vgs, which is relatively high with that JFet, can be 2.5 to 4V.
Thanks. What difference will I notice by eliminating R2 or increasing it to 4M or 6M. I value your advice - I'm just trying to understand.
- Status
- Not open for further replies.